This invention relates to categorizing heart disease, and more particularly to a method, and system for categorizing heart disease states according to anatomical shape and motion of the heart.
Cardiac disease categorization is a complex task, partly due to the fact that many disease states have terms that indicate where in the heart they occur. For example, “antereoseptal hypokinesis” is a diseased state where the myocardium is hypokinetic near the part of the septum, closest to the patient's chest. In order to categorize this information, highly detailed analytics may be performed with anatomy-location-aware information. Such information can be produced by direct comparisons between anatomical segmentations.
Known cardiac disease categorization methods based on statistical segmentation models can be problematic because of the need for training data for every single diseased, state to be segmented. Other methods extract segmentation-free feature descriptors, but these methods often do not provide a complete description of the anatomy and have trouble with anatomy-location disease categorization.